We identified, cloned and mapped a novel human gene, BRMS1 (Breast Metastasis Suppressor-1) which suppresses metastasis when transfected into two unrelated human breast carcinomas. Yet, tumorigenicity was not suppressed in the transfectants. These characteristics define a metastasis-suppressor gene. The novel BRMS1 gene maps to 11q13.1-q13.2, a chromosomal region commonly altered in late-stage, metastatic breast carcinoma. Because of its relatively small size, nuclear localization, presence of coiled-coil and leucine zipper motifs, we hypothesize that BRMS1 is part of a multimeric protein complex involved in transcriptional regulation. The objective of the current proposal is to determine the mechanism by which BRMS1 suppresses breast carcinoma metastasis. For Specific Aim 1, the underlying hypothesis is that precise disruption of the BRMS1 protein will result in the inability to suppress the metastatic phenotype. Following standard biochemical characterization (subcellular localization, post-translational modification), BRMS1 mutants will be generated by deletion and site-directed mutagenesis, transfected into metastatic human breast carcinoma cells (which do not express endogenous BRMS1) and evaluated for the mutants' effects o metastasis. Initial studies will disrupt the putative nuclear localization signals to determine whether nuclear localization is critical for BRMS1 function. Introduction of mutations to delete or modify specific domains will then be done to assess which are critical. Subsequently, missense mutations (e.g., alanine scan of critical amino acids) will introduce more defined mutations. In a complementary study, BRMS1-interacting proteins will be identified by co-precipitation and yeast two-hybrid analyses. Mutations which result in disruption of protein-protein interactions and metastatic potential will identify the critical amino acids responsible. Using this combined approach, a structure-activity relationship can be developed for BRMS1. Specific Aim 2 will generate homozygous null brms1 mice. This aim will test whether genetic loss of brms1 will result in increased metastasis by primary murine mammary carcinomas. A conditional knockout approach using Cre-Lox will be used in order to overcome the complications associated with a lethal mutation. Introduction of Lox-flanked brms1 into the germline will be done, followed by recombination using Cre under control of the MMTV and WAP promoters (giving wisespread and mammary-tissue-specific deletion, respectively). brms1-null mice will be bred to transgenic mice which routinely develop mammary tumors and the incidence and frequency of metastasis determined. The hypothesis is that mice lacking brms1 will develop more metastasis in more sites than mice which still express brms1.